The card describes a comprehensive evaluation across 12 industry-standard benchmarks. Aggregate construction, metric, sampling, scaffold and thinking mode are not stated.
条件
GLM-4.5目标列;The card describes a comprehensive evaluation across 12 industry-standard benchmarks. Aggregate construction, metric, sampling, scaffold and thinking mode are not stated. 安全、内部子项或跨测试聚合仅保留原始披露,不进入常规展示。
GLM-4.5 · Reasoning · GLM-4.5模型卡;Reasoning Benchmarks: MMLU-Pro, AIME 24, MATH 500, SciCode, GPQA, HLE, LCB (2407-2501) · Scores and configurations
GLM-4.5 record for Reasoning · GLM-4.5模型卡;Reasoning Benchmarks: MMLU-Pro, AIME 24, MATH 500, SciCode, GPQA, HLE, LCB (2407-2501). 1 score record for this release, including 1 exact value (may include historical records). Includes 1 source location.
The card describes a comprehensive evaluation across 12 industry-standard benchmarks. Aggregate construction, metric, sampling, scaffold and thinking mode are not stated.
Conditions
GLM-4.5目标列;The card describes a comprehensive evaluation across 12 industry-standard benchmarks. Aggregate construction, metric, sampling, scaffold and thinking mode are not stated. 安全、内部子项或跨测试聚合仅保留原始披露,不进入常规展示。